Date: Wed, 28 Aug 2013 09:01:38 +0000 (UTC) From: Guido Falsi <madpilot@FreeBSD.org> To: ports-committers@freebsd.org, svn-ports-all@freebsd.org, svn-ports-head@freebsd.org Subject: svn commit: r325515 - head/graphics/amide/files Message-ID: <201308280901.r7S91c48074575@svn.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: madpilot Date: Wed Aug 28 09:01:38 2013 New Revision: 325515 URL: http://svnweb.freebsd.org/changeset/ports/325515 Log: Fix build on recent current with clang. Modified: head/graphics/amide/files/patch-src-alignment_mutual_information.c (contents, props changed) Modified: head/graphics/amide/files/patch-src-alignment_mutual_information.c ============================================================================== --- head/graphics/amide/files/patch-src-alignment_mutual_information.c Wed Aug 28 08:26:22 2013 (r325514) +++ head/graphics/amide/files/patch-src-alignment_mutual_information.c Wed Aug 28 09:01:38 2013 (r325515) @@ -1,5 +1,5 @@ ---- src/alignment_mutual_information.c.orig 2011-07-15 07:08:17.000000000 +0200 -+++ src/alignment_mutual_information.c 2011-09-19 14:55:39.733328817 +0200 +--- src/alignment_mutual_information.c.orig 2012-01-22 17:25:06.000000000 +0100 ++++ src/alignment_mutual_information.c 2013-08-28 10:30:30.265534615 +0200 @@ -28,6 +28,13 @@ #include "amitk_data_set_DOUBLE_0D_SCALING.h" #include "alignment_mutual_information.h" @@ -14,3 +14,12 @@ /* this algorithm will calculate the amount of mutual information between two data sets in their current orientations */ /* it is a re-write of the original algorithm for purposes of improved speed. the hope is that it won't affect accuracy. */ /* rather than computing mutual information for the whole volume of data, the algorithm computes it for three orthogonal */ +@@ -209,7 +216,7 @@ + //g_print("\t\%i", mutual_information_array[i][j] ); // for point-wise counts + // g_print("\t\%4.3f", incremental_mi ); // for point-wise probability + +- if isinf(incremental_mi) { ++ if(isinf(incremental_mi)) { + //count it (because lots and lots of zeroes mean bad registration) and go to the next loop + mi_nan_count++; + // this is most often a problem when the incemental MI is infinity, due to divide by zero
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201308280901.r7S91c48074575>